The emergence of hyperthermia as a cancer therapy has been an exciting clinical development even though its full potential is still unknown. One of the main problems has been correlating the observed tumor response with the temperature distributions achieved during treatment. Lack of knowledge of complete SAR and temperature distributions within the target tissue volume has led to confusing and conflicting results which will ultimately be used to draw conclusions about hyperthermia as a clinical cancer therapy if the gap in existing knowledge is not closed. The principal hypothesis of the proposed work for addressing this clinical need is that the utilization of treatment measurements in conjunction with computational models as an integrated prognostic tool is of significantly higher clinical value than either measurement or computation used alone. To this end, electrical inverse profiling techniques will be used to estimate both SAR and temperature distributions. In the latter case, differential changes in reconstructed complex permittivity profiles will be correlated to temperature rise through empirical relationships and directly measured temperature data. The primary strengths of this approach are that the underlying physics are well-represented mathematically by the Maxwell equations, changes in the complex permittivities of tissues have established correlations with physiological effects of interest in hyperthermia, and the required measurement techniques and hardware are largely available. The proposed techniques will also incorporate invasive measurements -- a major advantage over completely noninvasive imaging and temperature sensing technologies where electrical inverse profiling has been used. Since clinical hyperthermia treatments already involve invasive measurements and some even utilize invasive heating sources, the demand for internal measurements does not represent a significant departure from the norm. However, when coupled with electrical inverse profiling, internal measurements may provide the information necessary to overcome limitations in noninvasive methodologies, thus offering a technique for accurately estimating complete SAR and temperature distributions during hyperthermia. With knowledge of complete temperature distributions, relationships between therapeutic outcome and variation in tumor temperatures can be determined. Then, the sensitivity of the desired temperature profile to controllable treatment parameters can be systematically studied and treatment protocols which maintain a high likelihood of achieving the desired response can be developed.